Statistical stability is how well the results of your study or experiment hold up.
More specifically, it’s a measure of how well you control for random errors in your study.
Statistical stability can be defined more precisely for specific fields. For example, let’s say you’re working with signal-to-noise ratio. Josselin Garnier and George Papanicolaou, in the book Passive Imaging with Ambient Noise, describe it as meaning
“…high signal-to-noise ratio of the quantity considered, including the image itself.”
How Do I Make Sure My Results Have Statistical Stability?
Ways to ensure statistical stability to test your null hypothesis include:
- p-values. A p value is used in hypothesis testing to support or reject the null hypothesis. It is the evidence against a null hypothesis. In general, the smaller the p-value the better.
- confidence intervals. For example, you might report a 95% confidence interval with your results.
Aschengrau, A. & Seage, G. Essentials of Epidemiology in Public Health.
Josselin Garnier and George Papanicolaou. Passive Imaging with Ambient Noise,Retrieved September 18, 2019 from: https://books.google.com/books?id=9jrzCwAAQBAJ